Page 1 of 2

Upgraded Nagios XI to 5.6.2 and recevived many false alerts

Posted: Mon May 20, 2019 8:02 am
by MOHANREDDY
Hi,

Last Friday we upgraded Nagios XI from 5.6.1 to 5.6.2 and the installation is successful. but after 5 mins users are bombarded with lots of email alerts saying Return code of 13 for service / Disk Usage on host was out of bounds , All the alerts says out of bounds, we rolled back to previous version(5.6.1) from snapshot. The alerts were no more and no notifications nothing, everything is fine.

I would to like to know the reason for those alerts and what do they mean?

Re: Upgraded Nagios XI to 5.6.2 and recevived many false ale

Posted: Mon May 20, 2019 12:30 pm
by scottwilkerson
What type of checks were these "/ Disk Usage" ?

Unfortunately we may need some information when updated to 5.6.2 to debug the problem such as permissions in this directory

Code: Select all

ls -al /usr/local/nagios/libexec/
What I'm afraid of is that you have a custom umask or something that is changing the permissions of the files when changed.

Re: Upgraded Nagios XI to 5.6.2 and recevived many false ale

Posted: Tue May 21, 2019 8:39 am
by smblair
We had the same experience yesterday but were able to revert the RPMS and not restore a snapshot. Our update was from 5.6.1 to 5.6.2. Running database recovery had no effect. Immediately after downgrading, the system stabilized and started sending recovery alerts.

Re: Upgraded Nagios XI to 5.6.2 and recevived many false ale

Posted: Tue May 21, 2019 1:24 pm
by MOHANREDDY
scottwilkerson wrote:What type of checks were these "/ Disk Usage" ?

Unfortunately we may need some information when updated to 5.6.2 to debug the problem such as permissions in this directory

Code: Select all

ls -al /usr/local/nagios/libexec/
What I'm afraid of is that you have a custom umask or something that is changing the permissions of the files when changed.
Currently, we have a default mask settings as below,
]# umask
0022
we didn't change any mask before or upgrade, a small correction I did restore from the backups not from snapshot. Today now again I upgraded on another server from 5.6.1 to 5.6.2, it again started alerting a lot of emails.

see the attached doc of the output you asked.

Thanks,

Re: Upgraded Nagios XI to 5.6.2 and recevived many false ale

Posted: Tue May 21, 2019 1:33 pm
by smblair
Here is our listing.

Re: Upgraded Nagios XI to 5.6.2 and recevived many false ale

Posted: Tue May 21, 2019 2:19 pm
by scottwilkerson
I'm confused, you showed 2 different things
the libexec_files.docx has completely wrong permissions

lib-perms.gz has permissions that should work fine..

Are you working together on the same server?

Re: Upgraded Nagios XI to 5.6.2 and recevived many false ale

Posted: Tue May 21, 2019 2:21 pm
by smblair
scottwilkerson wrote:I'm confused, you showed 2 different things
the libexec_files.docx has completely wrong permissions

lib-perms.gz has permissions that should work fine..

Are you working together on the same server?
No, no. Completely separate. They were just having the same issue we were.
Would you prefer our issue in a separate post? We do have paid support.

Re: Upgraded Nagios XI to 5.6.2 and recevived many false ale

Posted: Tue May 21, 2019 2:38 pm
by scottwilkerson
smblair wrote:No, no. Completely separate. They were just having the same issue we were.
Would you prefer our issue in a separate post? We do have paid support.
Yes please

Re: Upgraded Nagios XI to 5.6.2 and recevived many false ale

Posted: Tue May 21, 2019 2:41 pm
by scottwilkerson
MOHANREDDY wrote:
scottwilkerson wrote:What type of checks were these "/ Disk Usage" ?

Unfortunately we may need some information when updated to 5.6.2 to debug the problem such as permissions in this directory

Code: Select all

ls -al /usr/local/nagios/libexec/
What I'm afraid of is that you have a custom umask or something that is changing the permissions of the files when changed.
Currently, we have a default mask settings as below,
]# umask
0022
we didn't change any mask before or upgrade, a small correction I did restore from the backups not from snapshot. Today now again I upgraded on another server from 5.6.1 to 5.6.2, it again started alerting a lot of emails.

see the attached doc of the output you asked.

Thanks,
Somehow your permissions are changing on the libexec directory

Run the following after upgrade if they still look the same:

Code: Select all

chmod ug+x -R /usr/local/nagios/libexec/
chmod u+s -R /usr/local/nagios/libexec/check_dhcp
chmod u+s -R /usr/local/nagios/libexec/check_icmp

Re: Upgraded Nagios XI to 5.6.2 and recevived many false ale

Posted: Tue May 21, 2019 3:22 pm
by MOHANREDDY
why does permissions change after each and every release, we have such type of issues when we upgraded from 5.5.11 to 5.6.1 and again the same repeated from 5.6.1 to 5.6.2?